SOFT COMPUTING APPLICATION IN SOIL CLASSIFICATION ANALYSIS

Abstract eng:
This paper deals with a new application of soft computing methods in soil classification procedure that is technically easy to use and does not require complex mathematical definitions for the model components. Two types of radial-basis neural networks, namely, probabilistic neural networks and generalized regression neural networks were used in this study. The proposed quick-automatic soil classification methods are implemented to analysis number of ground-motion stations located at North-West of Iran. The efficiency of proposed methodologies is compared with soil classification results based on measurement of the average shear wave velocity at a subsurface depth of 30 m. In addition, the results from neural network-based soil classification system were evaluated and compared with two empirical schemes which are based on the spectral shape of normalized response spectra and the average horizontal to vertical spectral rations of ground motions. The results revealed the acceptable ability of proposed methods to predict the soil class in the study area.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket